INDEX

Explanations

terms and phrases indicating positive outcomes or successful actions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.MixedReality

-0.16

mj

-0.15

 Champion

-0.15

 trick

-0.15

elman

-0.15

setMessage

-0.14

ëĤ¨

-0.14

aml

-0.14

oi

-0.14

iny

-0.14

POSITIVE LOGITS

 komun

0.16

áº§m

0.15

ulta

0.15

0.14

IBE

0.14

 cá»Ļng

0.14

ings

0.14

ibe

0.14

acre

0.14

istics

0.13

Activations Density 0.026%

terms and phrases indicating positive outcomes or successful actions

No Comments

No Known Activations